June is, 1921 
Life-History Studies of Three Jointworm Parasites 407 
purpose (PI. 74, B). This cage was made by cutting window glass to 
the size of ordinary glass slides, and by the use of a dentist’s small burr 
or carborundum grinder making a single small cell large enough to accom¬ 
modate the larva of H. tritici in one surface of each slide. After a single 
egg with its host had been placed in each cell, the cell was closed with 
the ordinary glass cover slip, to which was applied a droplet of honey 
to hold it in place. The usual label was pasted on each slide for identi¬ 
fication, and the slides were then placed in small closed pasteboard 
boxes in order to exclude the light. The life stages of the parasites 
from egg to adult could thus be observed under the microscope. Often, 
however, the parasite larva would crawl away from its host and starve 
during the night. Sometimes the host larva would turn dark as if 
decaying and another would have to be substituted. After long expe¬ 
rience and with all possible care that could be given, it was found impos¬ 
sible to rear more than about 60 per cent of a given number of parasites 
by the cell-slide method. But since this method was the only one known 
in which the different stages could be observed continuously, it was 
adhered to throughout. Eupelmus allynii bred very freely by this 
method, while Homoporus chalcidiphagus bred less freely and Ditropinotus 
aureoviridis with some difficulty. The larvae of the last-named species 
completed their development very readily, but very few continued through 
to the adult stage. 
In order to have a control on the period of development of the parasites 
in the cell slides (PI. 79, B), some of the eggs of the same age were not 
removed from the Harmolita cells but were permitted to remain and 
develop normally as they would in the field. It was found that each of 
these species very closely approximated the same period of development 
when reared in the glass cell as when reared in the galls of the jointworm 
under the same weather conditions. 
In observing the larval development of these parasites it was com¬ 
paratively easy to identify the various instars of Ditropinotus aureoviridis 
and Eupelmus allynii by the length, number, and position of the setse 
and the change in size and shape of the larvae. The instars were verified 
by making a balsam mount of each cast skin. It was found impossible to 
identify all of the instars of Homoporus chalcidiphagus except by remov¬ 
ing to a balsam mount the entire contents of a glass cell, where the 
individual larva had developed, and noting the number of pairs of man¬ 
dibles remaining therein. 
All observations indicate that only one specimen of any of the three 
species studied ever matures in a single Harmolita cell. An individual 
host larva seems to furnish just enough food for a single parasite. Usually 
only the old dried skin of the host remains after the parasite larva has 
become full grown. 
